The traditional IP network does not support the terminal mobility, and in the related art, all kinds of different solutions are proposed, mainly including the mobile IPv4, the mobile IPV6, the agent mobile IPV6 of the IETF, the GPRS Tunneling Protocol (abbreviated as GTP) mode of the 3GPP, etc. This technology supports the terminal mobility by adopting the mode of the fixed anchor point, such as, the GPRS supporting node (GGSN) is stipulated in the WCDMA as the terminal mobile anchor; the mobile IP protocol is adopted in the CDMA network, and a local agent (HA) is treated as the anchor point. But the fixed anchor point will bring the problem that the data packet route is indirect, thus increasing the transmission delay and the bandwidth waste. With the development of the mobile wireless packet technology, such as, 2G/3G/4G, etc. the number of users and the flow rate in the mobile Internet are increased gradually, and a gateway device, such as, a GGSN, etc., shifts downward gradually, which will make the problem of the indirect route more serious.
The technology 1 correlated to the present document:
the patent application 1 of which the application number is 200910174823.5: “Method for obtaining/notifying information, forwarding and switching data message, and an access node”.
The above-mentioned patent application 1 proposes a switching management method and data message forwarding method under the Location-ID separation architecture based on the network, which solves the indirect route problem in the communication process of the mobile terminal data. What is involved during the switching is that:
(1). after the location of a mobile node is changed, the access gateway to which the mobile node is switched allocates a new location identification for the mobile node, and the mobile node will communicate with the correspondent node by using the new location identification;
(2). in order to ensure an optimal data forwarding path between the mobile node and the correspondent node, the access gateway to which the mobile node is switched initiates a procedure for updating the correspondent node mapping to realize the route optimization;
(3). in a procedure of updating the correspondent node mapping, the access gateway needs to obtain the correspondent node information, to determine which correspondent nodes it initiates the procedure of updating the correspondent node mapping to;
(4). in the method disclosed in the above-mentioned patent application, the access gateway mainly obtains the correspondent node information by monitoring and detecting the data flow, and maintaining information of the correspondent node which is performing the data communication with the mobile node.
The main problems in the patent application 1 are as follows:
(1). in order to acquire the correspondent node information, the access gateway needs to monitor and detect the data flow to judge the existence condition of the connection of the end-to-end host to maintain information of the correspondent node which is performing the communication with the mobile node. That process has consumed a large amount of handling capacity of the access gateways.
(2). the access gateway defines the aging mode of the correspondent node table, and the basic principle is to monitor the data flow of both ends of the communication; if there are data flows forwarded in a certain time period T1, then it means that the communication with the correspondent node is on-going; otherwise, the correspondent node in the correspondent node table is aged. Wherein, 1) if the both ends of the communication keep the communication connection, but there is no data message sent in a certain time period T2, and T2>T1, then the access gateway ages the connection of the both ends of the communication at this moment; 2) if the set value of T1 is too large, then after the service connection of the both ends of the communication is disconnected, the longest time required by the access gateway to age the connection of the both ends of the communication is T1. In the above-mentioned case, the correspondent node table of the access gateway cannot reflect the situation of the end-to-end communication connection in real time.
The technology 2 correlated to the present document:
the Location-ID Separation Protocol (abbreviated as LISP) adopts the route technology, and changes the current route topology structure to some extent, utilizes the minimum transformation to optimize the current route transport technology by combining the current transport network.
The host uses the IP address (called the end identification EID in the LISP system) to track socket connections, send and receive data packets.
The router transmits the data packets based on the IP destination addresses (route locations RLOCs).
The tunnel routers are introduced into the LISP system, the encapsulation is performed while initiating the host packet, and the data packet is de-capsulated before transmitting to the destination place finally. In an LISP data packet, the IP address of “the outer layer header” is RLOC. In the procedure of performing the end-to-end packet switching between the hosts of two networks, the ITR encapsulates each packet with a new LISP head, and strips the new head at the egress channel route. The ITR performs the EID-to-RLOC searching to determine the route path to the ETR, and the ETR regards the RLOC as its one address.
The proposition of the LISP is not to solve the mobility problem, while mainly solves the problem of the network size and aims at the problem that the route table is too big; for the problem that the mobility and the multi-home are solved incidentally after location-ID separation, there is no specific scheme and implementation method now.
The method of an access gateway monitoring, detecting and maintaining the correspondent node information defined in the solution of the patent application 1 consumes a large amount of processing capacity of the access gateway, and the real time performance is bad; the solution of realizing the route optimization during the switching is not found in the related art 2.